This repo is a collection of models, scripts, and Jupyter notebooks that I experiment with as I learn more about machine learning. Some of the models and helper functions found in here have been copied from other sources and in those cases I provide a link to the original work. My intention is to initially copy those models and tweak them to better understand how they work and along the way create my own models.
My goal is to explore recent advancements in NLP/NLG, image recognition, world modeling, and reinforcement learning.
Links to blogs and whitepapers I read and use as inspiration:
- World Models
- MuZero
- MMD Variational Autoencoders
- Attention Is All You Need
- An Image is Worth 16x16 Words
pipenv shell && pip install
Below is a collection of datasets I use when training these models.